Brief · Use case · Innovation
Engagement Drop Diagnosis.
LiveDiagnosisFind out why patients stop using a digital therapeutic after the first month, and which explanation the usage data support.
Engagement Drop Diagnosis is a ready-made use case for Innovation. It ships under the Diagnosis template.
For digital health product leads watching retention fall. You get where the drop is largest, onboarding, content fit, access and population explanations weighed against the evidence, the likely cause with a confidence level and the checks that would settle it. Claims are cited or marked as gaps.
Add it to your library to reuse from New → From template. Edit the saved copy in your library on the service form.
The decision
Why does patient engagement with a prescription digital therapeutic fall sharply after the first month of use: onboarding friction, a mismatch between content and patient need, device or access problems, or a change in who is being prescribed the product?
The explanations weighed
The diagnosis restates the fall in measurable terms first: the share of users active by week, where the drop is largest, and for which groups. It sets that against what is known about drop-out from app-based interventions.
| Explanation | What it would look like | What would confirm or refute it |
|---|---|---|
| Onboarding friction | Users cannot get started or do not see the point in the first sessions, and leave early. | Confirmed by drop-off concentrated in the first sessions and by setup steps with high failure. Refuted if early use is steady and the drop comes later. |
| A mismatch between content and need | The content does not fit what the patient needs, so use fades once novelty ends. | Confirmed by a drop that varies by patient segment and by module. Refuted if every segment drops alike. |
| Device or access problems | Users cannot use the product reliably because of a device, a connection or an account problem. | Confirmed by technical logs and support tickets that match the drop. Refuted if logs are clean. |
| A change in who is prescribed the product | The mix of patients changed, for example after a wider prescribing base, and the new patients engage less. | Confirmed by a cohort comparison in which earlier and later cohorts differ at the same week. Refuted if cohorts match. |
What the diagnosis does
A drop is expected; the question is how much. A systematic review of app-based interventions for chronic disease found a pooled dropout rate of 43 percent, with a wide range between studies and higher rates in observational studies than in randomized trials. The diagnosis compares the product’s curve with that range, and does not treat attrition as a fault in itself.1
Cohorts separate the explanations. A fall that appears at the same week in every cohort points to the product. A fall that differs between cohorts points to who is being prescribed. The diagnosis compares cohorts before it names a cause.
What the Alliance means by a digital therapeutic. The Digital Therapeutics Alliance describes these products as clinically evaluated software that delivers medical treatment to patients, in three categories: to treat, manage or prevent a disease. The diagnosis checks whether the engagement measure used suits the product’s category.2
How it could be wrong. The likely cause rests on an assumption, such as that engagement is measured the same way across versions. The diagnosis names it and the check that would break it.
Recommended length
Comprehensive
The question asks you to weigh at least three explanations. The matrix that does that, and the Pareto view of where the effect sits, are in the Comprehensive length. Standard names a cause without the side-by-side.
You can pick another length on the form. The other two stay available.
Assumptions the diagnosis checks
The diagnosis states the assumptions its likely cause rests on and sets what would break each one. Examples for this problem:
- Usage data are measured the same way across app versions and cohorts.
- Engagement, as defined, reflects the dose of treatment the label requires.
- The support and technical logs are complete.
Standards, guidance and sources
Attrition and context
The decision practice behind the memo itself is on the template page.
What gets flagged
Any engagement measure that changed between versions, any cohort too small to compare, and any explanation the usage data cannot separate.
What arrives
- The problem as stated
- What is actually happening
- Competing explanations
- The likely cause
- What would confirm it
- Where the effect concentrates
- What to rule out
What it will not do
It does not decide what to do about the cause. A Decision Memo takes the diagnosis as its starting point.
It does not redesign the product, and it does not say how much engagement the label requires.
It does not prove the cause. It names the most likely cause, says how sure it is, and says what would show it to be wrong.
It does not assign blame. It looks for contributing causes in the system, in the way a blameless postmortem does.
It does not act. Deciding what to do about the cause is a different question; a Decision Memo is the template for it.
References
- [1] Meyerowitz-Katz, G. et al. Rates of attrition and dropout in app-based interventions for chronic disease: systematic review and meta-analysis. Journal of Medical Internet Research, 2020;22(9):e20283. PMC7556375. Pooled dropout 43 percent (95 percent confidence interval 29 to 57), as given in the paper’s abstract by search excerpts; 17 studies.
- [2] Digital Therapeutics Alliance. Understanding DTx. Read on 9 October 2026.
Links checked 9 October 2026
These are the sources the template drew on. None of their authors or issuers has reviewed it.
About the Brief service
A Brief is a written answer from one specialist, at the depth you asked for. It arrives with its sources, and it closes by saying plainly what would change the conclusion. Nothing is hidden behind a summary, and anything the expert could not source is marked as a gap rather than filled in.
To edit a saved template, open New → Brief → From template, or the library rail on the service form.